201. Improved non-rigid medical image registration algorithm.
- Author
-
ZHANG Jing-ya and WANG Jia-jun
- Abstract
The non-rigid registration is one of the most important research areas for various medical image analysis tasks and applications. This paper proposed a novel non-rigid image registration algorithm based on the traditional local afine transformation model. In order to protect the image local detaiis and edges, it used an adaptive nonlinear regularization term which took into account the regional statistical intensity information and the magnitude characteristic of the Chebyshev low pass filter. In the framework of the multi-resolution layered refining and coarse-to-fine warping strategy, the proposed algorithm achieved good performance in the registration of mono-modality medical images with relative larger deformations and the fine registration of multi-modality images with intensity variations which couldn't be processed by traditional algorithms. The experimental results show that the proposed algorithm is effective for non-rigid medical image registration. [ABSTRACT FROM AUTHOR]
- Published
- 2015
- Full Text
- View/download PDF